Experiments

Experimentation and Research

• We tend to believe, after learning an outcome, that we would have foreseen it - the “I knew it all along” phenomenon.

• Overconfidence: we tend to think that we know more than we do.

  • Overconfidence will inhibit critical thinking or creativity.

• Critical Thinking

  • Thinking that does not blindly accept arguments and conclusions

  • Examine assumptions, find hidden values, evaluate evidence

  • OVERCONFIDENCE WILL INHIBIT THIS

Parts of Research

• Theory

  • An explanation using an integrated set of principles that organizes and predicts observations.

• Hypothesis

  • A testable prediction, often implied by a theory.

• Operational Definition

  • A statement of procedures (operations) used to define research variables.

• Replication

  • It repeats an experiment in order to compare results and figure out that your actual experiment was the correct answer, as well as whether the basic finding generalizes to other participants and circumstances.

• Experimentation

  • An investigator manipulates one or more variables to observe their effect (i.e. on a mental process or behavior - a dependent variable

  • Done by random assignment of participants the experiment controls by other relevant factors

  • For a study to be considered an experiment, it needs to include either random assignment or manipulation of one independent variable

Independent Variable

• Whatever is being manipulated in an experiment, which hopefully brings about change.

  • If there is a drug in the experiment, it is almost always the independent variable.

Dependent Variable

• Whatever is being measured in the experiment, dependent on the IV.

  • The dependent variable would be the effect of the drug.

Confounding Variables

• Anything that can influence or compromise the results of the experiment are the confounding variables.

  • The object of the experiment is to prove that A causes B!

Descriptive Studies

Descriptive Studies

• Psychologists describe behavior using three different methods:

  • Case studies

  • Surveys

  • Naturalistic observation

Case Study

• They study one or more individuals in depth in hope of gaining information and learning things that hold true about us all

Survey

• A technique used for finding the behavior or attitude of people, which is self reported

• Done by questioning a representative, random sample of people

  • A random sample is a sample that accurately represents a population since each member has an equal chance of being included in the sample.

Naturalistic Observation

• Observing and recording behavior in settings that are natural to the subjects without trying to change anything about the circumstances

Phenomena, Bias, and Other Effects

Hawthorne Effect

• There is no control group

  • Selecting a group of subjects to experiment with affects the performance, regardless of what the experiment is

  • Knowing they are in an experiment will change their performance either way

False Consensus Effect

• We have a tendency to overestimate how much others share our beliefs or thoughts

Placebo Effect

• The belief that a person is on the real drug, and even though they aren’t truly on the drug, experience the effects that are from the active agent.

• Can act as a confounding variable

• Works using the belief of positive expectations

Experimenter Bias

• A confounding variable that is not consciously controlled

A double-blind procedure can stop any experimenter bias

Parts of a Study

Population

• All the cases in a group from which samples are drawn from during a study or similar experiment

Samples

• There are two types of examples

  • Stratified: a population is divided into categories from which a random sample is taken

  • Random: Everyone in a population has an equal likelihood of being selected for a study

Experimentation

• There are two conditions to experimentation

  • Experimental condition: the condition that exposes participants to one version of the independent variable

  • Control condition: the condition that contracts with the experimental condition, and serves as a comparison to evaluate the experiment’s effects.

• Random assignment

  • Assigning participants to experimental and control groups by chance

  • Minimizes pre-existing differences between the people assigned to different groups.

• Placebo

  • An inert substance or condition that is put into an experiment that is administered in place of an active agent, with the intent to see if it triggers the effect that characterizes the active agent.

• Blind and Double-Blind Procedures

  • Blind procedures mean that the participants are unaware of which group they are in - experimental or control

  • Double-blind procedures mean that nobody, neither the participants or research staff, know which participants are in which groups, which is commonly used in drug evaluation studies.

Correlation Studies

Correlational Study

• Examines relationships between two variables

  • Correlation does NOT prove causation

• Has a correlation coefficient from -1.0 to +1.0

  • A statistical measure of the extent to which two factors vary together, and thus how well either factor predicts the other

  • Defines the strength of a correlation

  • Positive means one variable increases with the other

  • Negative means as one increases, the other decreases

• Positive vs. negative correlation

  • Positive correlation means that the variables go in the same direction

  • Negative correlation means that the variables go in opposite directions

Scatter Plot

• A graphed cluster of dots, each of which represents the values of two variables

  • The slope of the points suggests the direction of the relationship

  • The amount of scatter suggests the strength of the correlation

    • little scatter indicates high correlation

• Also called a scattergram or scatter diagram

Interpreting Data

Measures of Central Tendency

• Ways to determine parts of data, including those of variables that occur most frequently, those that are in the middle of a dataset, and the average of the results.

• Mean, median, and mode

Approving an Experiment

• An experiment must be approved before it can be conducted

  • It has to be reviewed by the ethics board

• The ethical guidelines are established by the American Psychology Association (APA).

P-Value

• A hypothesis test that is used to determine the significance of the results from a study.

• The probability that results from an experiment are due to chance and not the experimental conditions.

• The null hypothesis is a prediction or claim about a population and hypothesis test

• P-value is used to see how valid the null hypothesis claim is.

• Range from 0 to 1

  • P value of less than .05 is statistically significant, as it had a less than .05% chance that it was chance and not the experimental conditions.